Improved two-vector model predictive torque control for PMSM

Abstract

Aiming at the problems of the traditional model predictive control (MPTC) of permanent magnet synchronous motor (PMSM) that the weight coefficient setting is cumbersome and the torque and flux pulsation are large, an improved PMSM two-vector model is proposed to predict the torque. This method expands the virtual vector to increase the voltage candidate vector, combines the voltage vector to act on the torque and flux linkage principle, selects the voltage vector, and reduces the calculation amount of the candidate vector. The evaluation mechanism of the selected vector is set by combined with the duty cycle method, so the weight factor of cost function in the traditional method is eliminated. The simulation results show that the improved method proposed can not only maintain high dynamic performance, but also improve the control effect of torque and flux linkage in the steady state of the system.